๐Ÿงฐ Installation Guide โ€” SemiDevKit

This document explains the environment setup, required software, and initial installation steps
for using SemiDevKit.

Target tools:


๐Ÿ–ฅ 1. System Requirements

1.1 Supported OS

OS Status Notes
๐Ÿง Linux โœ… Recommended Ubuntu 20.04 / 22.04
๐ŸชŸ Windows โœ… Recommended WSL2 + Ubuntu
๐ŸŽ macOS โœ… Supported Intel / Apple Silicon

๐Ÿ’ก Using WSL2 is strongly recommended on Windows environments.


๐Ÿ“ฆ 2. Required Software

๐Ÿ 2.1 Python

SemiDevKit supports the following Python versions:

โœ” Check Python version

python3 --version

โœ” Install basic packages

pip install numpy scipy matplotlib pandas pyyaml

๐Ÿ“Œ Some modules run with only numpy and matplotlib,
but analysis tools (BSIM / Paramus) require pandas.


โšก 2.2 ngspice

SemiDevKit uses ngspice for SPICE-based simulations.

๐Ÿง Linux (Ubuntu)

sudo apt update
sudo apt install -y ngspice

๐ŸŽ macOS

brew install ngspice

๐ŸชŸ Windows

โœ” Verification

ngspice --version

๐ŸชŸ 3.1 WSL2 (Windows Only)

WSL2 is essential for stable operation on Windows.

wsl --install

After installing Ubuntu from the Microsoft Store:

sudo apt update && sudo apt upgrade -y

๐Ÿณ 3.2 Docker (for OpenLane-Lite)

Docker is required to run the minimal RTL โ†’ GDSII flow.

โœ” Required settings


๐Ÿ“ 3.3 Visual Studio Code

Recommended editor environment.


๐Ÿ“ฅ 4. Clone the Repository

๐Ÿ” HTTPS

git clone https://github.com/Samizo-AITL/SemiDevKit.git
cd SemiDevKit

๐Ÿ”‘ SSH

git clone git@github.com:Samizo-AITL/SemiDevKit.git
cd SemiDevKit

๐Ÿ 5. Python Environment (venv)

SemiDevKit consists of multiple independent tools.

๐Ÿ“Œ Using separate virtual environments per tool is recommended.


5.1 Create and activate venv

(Linux / WSL2 / macOS)

cd SemiDevKit
python3 -m venv .venv
source .venv/bin/activate

5.2 Create and activate venv

(Windows PowerShell)

cd SemiDevKit
python -m venv .venv
.\.venv\Scripts\Activate.ps1

5.3 Install Python packages

pip install -r requirements.txt

If requirements.txt is not available (e.g., early releases):

pip install numpy scipy matplotlib pandas pyyaml

๐Ÿงญ 6. ngspice Path Configuration (Windows Native)

setx PATH "$env:PATH;C:\Program Files\Spice64\bin"
ngspice

๐Ÿ“ 7. Directory Overview

SemiDevKit/
 โ”œ bsim/        # BSIM4 analyzers (DC / CV / DIM / Reliability / Paramus)
 โ”œ tcad/        # TCAD playgrounds (MOSFET / PZT)
 โ”œ openlane/    # OpenLane-Lite (RTL โ†’ GDSII)
 โ”œ docs/        # Documentation
 โ”œ assets/      # GitHub Pages assets
 โ”œ README.md
 โ”” ChangeLog.md

๐Ÿš€ 8. Quick Test

โœ” ngspice + BSIM Analyzer

cd bsim/analyzer_dc
python run/run_vgid.py

โœ” Python Plot Test

python - <<EOF
import numpy as np
import matplotlib.pyplot as plt
plt.plot([0,1],[0,1])
plt.savefig("test.png")
print("OK")
EOF

๐Ÿ›  9. Troubleshooting

โŒ ngspice not found

โŒ venv activation permission error (Windows)

Set-ExecutionPolicy -Scope CurrentUser RemoteSigned

๐Ÿ“œ 10. License Notes

SemiDevKit uses a hybrid license model.

Scope License
๐Ÿ’ป Source Code MIT License
๐Ÿ“˜ Documentation / Diagrams CC BY 4.0

๐Ÿ“ฌ 11. Contact

Item Details
๐Ÿ‘ค Name Shinichi Samizo
๐Ÿง‘โ€๐Ÿ’ป GitHub Samizo-AITL

๐ŸŽ‰ After installation, proceed to docs/UsageGuide